Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    ui interaction states: hover, active, focus

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    u
    0.70
    ol
    0.62
    E
    0.62
    j
    0.59
    n
    0.58
    Parse
    0.57
    re
    0.57
    ro
    0.57
     ["
    0.57
    U
    0.56
    POSITIVE LOGITS
     Calyce
    0.73
    드는
    0.61
    ᚄ
    0.58
    드가
    0.57
     chức
    0.56
     Cál
    0.56
    𝕕
    0.56
     dilute
    0.56
     дітей
    0.55
    スの
    0.55
    Act Density 0.007%

    No Known Activations